1. Field of the Invention
The present invention relates to a humanized antibody or an antibody fragment thereof, which specifically recognizes a polypeptide encoded by CD27 gene containing an O-linked sugar chain to which galactose is not bound, and binds to its extracellular region of the peptide; a hybridoma which produces the humanized antibody; a DNA which encodes the humanized antibody; a vector which comprises the DNA; a transformant obtainable by transforming the vector; a process for producing a humanized antibody or an antibody fragment thereof using the hybridoma or the transformant; and a diagnostic agent using the antibody or the antibody fragment thereof, or a therapeutic agent comprising the humanized antibody or the antibody fragment thereof as an active ingredient.
2. Brief Description of the Background Art
In recent years, there have been reported some cases in which the onset of various diseases or the progression of pathology is accompanied by structural changes in sugar chains attached to the protein which is expressed by cells involved in the disease or pathology thereof. Representative ones among these cases are an expression of Tn antigen which is one of the O-linked (serine/threonine type) sugar chain antigens and which is found in more than 80% of human cancer types, and an expression of a sialyl-Tn antigen in which sialic acid is attached to the Tn antigen (Non-Patent Literature 2).
It is known that the expressions of these sugar chain antigens are rarely found in normal cells, and research for applying them as target molecules of cancer-specific vaccine therapies to medical care has been carried out (Non-Patent Literature 1). The expressions of the cancer-specific sugar chain antigens are under the control of the activity of enzymes constituting the complicated biosynthetic pathway of sugar chains and the complicated metabolic pathway of sugar chains in living organisms. For example, it is known that, in cancer cells, changes in the expression pattern of a gene encoding the proteins responsible for the biosynthetic pathway of sugar chains lead to brockage on the biosynthetic pathway of sugar chains.
The Tn antigen is known as an intermediate of the biosynthetic pathway of an O-linked sugar chain in normal cells, and has a structure (GalNAc α-Ser/Thr) in which N-acetylgalactosamine (GalNAc) is α-bound to a hydroxyl group on a side chain of a certain serine (Ser) or threonine (Thr) residue of an amino acid sequence of a protein.
Biosynthesis of a normal-type O-linked sugar chain (such as TF antigen) takes place by the transfer of one molecule of galactose to the non-reducing terminal of the Tn antigen by the activity of core 1β3 galactosyltransferase (core 1β3Gal-T, T-synthetase). It is considered that the biosynthetic pathway of sugar chains is blocked as a result of decrease in the activity of intracellular core 1β3 galactosyltransferase, and thereby the Tn antigen or the sialyl-Tn antigen is expressed in many types of cancer cell lines.
The mechanism of the decrease in the activity of core 1β3 galactosyltransferase in cancer cells is complicated and has not yet been fully elucidated. However, as one possible mechanism, it has been supposed that the intracellular core 1β3 galactosyltransferase activity is greatly decreased due to a mutation in a gene encoding a certain chaperone protein (Cosmc) which is necessary for the activity expression of core 1β3 galactosyltransferase (Non-Patent Literature 6).
Based on the fact that expression of the Tn antigen is commonly found among plural cancer types, it is considered that aberration in the biosynthetic pathway of sugar chains or the metabolic pathway of sugar chains in cells is a main cause of common changes in structures of sugar chains attached to many different glycoproteins expressed in the cells.
Cancer is a representative disease which is known to have a close relationship between the structural change of a sugar chain and the progression of pathology. Other than cancer, IgA nephropathy is known as another disease which is known to have a close linkage between the structural change of a sugar chain and the pathological progression. IgA nephropathy is chronic glomerular nephritis which is pathologically characterized by showing granular deposition of one of the immune globulin, immunoglobulin A (IgA), in the glomerular mesangium, and was first reported by Berger in 1968 (Non-Patent Literature 2).
IgA nephropathy is representative nephritis accounting for about half of chronic glomerular nephritis patients in Japan. It is said that about 40% of patients who have been diagnosed with IgA nephropathy will undergo a transition of the disease to late-stage renal failure within 20 years, and who will inevitably receive hemodialysis, renal transplantation or the like. As described above, even though IgA nephropathy has been generally recognized as a poor-prognosis disease, a clinically-validated therapy has not yet been established.
There is known that IgA1, out of two different IgA isotypes (IgA1 and IgA2), is mainly deposited in the kidney in the bodies of patients of IgA nephropathy. In addition, as a cause of IgA1 deposition, it has been reported that a structure of an O-linked sugar chain attached to a hinge region present on the IgA1 molecule, but absent on the IgA2 molecule, changed from a normal type to a Tn or sialyl-Tn antigen (Non-Patent Literatures 3 and 4).
It was demonstrated that once the deficiency of galactose from a O-linked type sugar chain added to the IgA1 hinge region has resulted in conversion of the sugar chain into a Tn or sialyl-Tn antigen, self-agglutination ability of the IgA1 molecule is enhanced, and deposition of the IgA1 molecule into the renal mesangial areas is accelerated (Non-Patent Literature 5).
Further, a decline of the core 1β3 galactosyltransferase activity due to a decreased expression level of Cosmc has been reported in IgA-producing cells isolated from IgA nephropathy patients (Non-Patent Literature 6).
In other words, the biosynthetic pathway of sugar chains is blocked halfway through in IgA-producing cells in the bodies of IgA nephropathy patients and as a result, sugar chain-deficient IgA1 is produced instead of IgA1 having a normal type sugar chain. As one of the pathogenic mechanisms of IgA nephropathy, it is advocated that the inflammation is induced as a result of the deposition of this sugar chain-deficient IgA1 in the renal glomerulus.
Generally, IgA is produced by B cells in blood, or plasma cells differentiated from B cells. The plasma cell is the final stage of B-cell differentiation. The plasma cells are distributed in secondary lymphoid tissues, systemic mucosal tissues, bone marrow, etc., and produce large quantities of antibodies. It is known that IgA-producing plasma cells are distributed mainly in mucosal tissues.
On the other hand, it is known that, in the germinal center of secondary lymphoid tissues, memory B cells or plasma cells are differentiated from B cell clones which have acquired an ability to produce high-affinity IgA antibodies, and the thus differentiated cells are distributed throughout target organs in whole-body and continuously produce antibodies over an extended period of time.
However, it is unclear at which stage of the B cell differentiation process, the cells which produce the sugar chain-deficient IgA involved in the pathogenesis of IgA nephropathy are developed, and to which body tissues the B cells or plasma cells which produce the sugar chain-deficient IgA are distributed.
Among proteins known as a cell membrane surface molecule expressed in B cells or plasma cells, CD27 is known as one of the molecules to which an O-linked sugar chain binds (Non-Patent Literature 7). The CD27 molecule, belonging to a member of the tumor necrosis factor receptor (TNFR) superfamily, is a type I membrane protein having a molecular weight of about 55 kDa, and is present as a disulfide-linked dimer of two monomers (Non-Patent Literature 8).
It is known that CD27 is expressed in some of T lymphocytes as well as in plasma cells and B cells. In particular, it is known that an expression level of CD27 is elevated upon differentiation of B cells into memory B cells and plasma cells in the differentiation process of B cells. It is known that CD27 to which an O-linked sugar chain binds is expressed in these cells during the differentiation process, but an amino acid residue to which the sugar chain binds is not clearly demonstrated (Non-Patent Literature 9).
As a ligand molecule of CD27, CD70 belonging to the TNF family is known. It is known that CD70 binds to CD27 expressed in some of B or T cells, induces cell proliferation signals, and stimulates B cells to produce antibodies (Non-Patent Literature 10).
In addition, it is known that the expression of CD27 is enhanced in several types of cancer cells as well as in normal cells. As types of cancer expressing CD27, there have been reported a variety of non-Hodgkin lymphomas, such as mantle cell lymphoma, chronic lymphocytic leukemia, small lymphocytic leukemia, Burkitt's lymphoma, follicular lymphoma, MALT lymphoma, diffuse large B-cell lymphoma, plasmacytoma (Non-Patent Literature 11).
Many of cancer cells are known to express a sugar chain-deficient protein containing a sugar chain including Tn antigen, sialyl-Tn antigen and the like, as described above.
As an antibody which specifically recognizes CD27, there has been reported an S152 antibody obtained by immunizing leukemia cells isolated from patients with Sezary syndrome (Non-Patent Literature 8). However, the S152 antibody is also shown to have an affinity with normal B cells and T cells. Up to date, there has not been known such an antibody which specifically recognizes CD27 molecule containing an O-linked sugar chain to which galactose is not bound.
It is generally known that, when a non-human antibody such as a mouse antibody is administered to human, it is recognized as a foreign substance so that a human antibody for mouse antibody (Human Anti Mouse Antibody; HAMA) is induced in the human body. It is known that HAMA reacts with the administered mouse antibody to thereby induce side effects (Non-patent Literatures 12 to 15), enhances disappearance of the mouse antibody from the body (Non-patent Literatures 16 to 18) and decreases therapeutic effect of the mouse antibody (Non-patent Literatures 19 and 20).
In order to solve these problems, attempts have been made to prepare a human chimeric antibody or a humanized antibody from a non-human antibody using gene recombination techniques.
A humanized antibody has various advantages in administration to human in comparison with a non-human antibody such as a mouse antibody. For example, it has been reported that the immunogenicity was decreased and the blood half-life was prolonged in a test using monkey, in comparison with a mouse antibody (Non-patent Literatures 21 and 22). That is, the humanized antibody is expected to cause fewer side effects in human than non-human antibodies and have sustained therapeutic effect for a long time.
Also, since a humanized antibody is prepared using gene recombination techniques, it can be prepared as various forms of molecules. For example, when γ1 subclass is used as a heavy chain (hereinafter referred to as “H chain”) constant region (hereinafter referred to as “C region”) of a human antibody (H chain C region is referred to as “CH”), a humanized antibody having high effector functions such as antibody-dependent cellular cytotoxicity (hereinafter referred to as “ADCC activity”) can be prepared (Non-patent Literature 23), and prolongation of the blood half life in comparison with mouse antibodies can be expected (Non-patent Literature 24).
Particularly, in the case of treatment for removal of CD27-positive cells, cytotoxic activities such as complement-dependent cytotoxicity (hereinafter referred to as “CDC activity”) and ADCC activity via the Fc region (the region after the antibody heavy chain hinge region) of an antibody are important, in order to specifically damage the target cells by accumulating effector cells near a tumor tissue via the antibody. In the treatment of humans, a human chimeric antibody, a humanized antibody or a human antibody is preferably used for exhibiting the cytotoxic activities (Non-patent Literatures 25 and 26).
In addition, with recent advance in protein engineering and genetic engineering, the humanized antibody can also be prepared as an antibody fragment having small molecular weight, such as Fab, Fab′, F(ab′)2, a single chain antibody (hereinafter referred to as “scFv”) (Non-patent Literature 27), a dimerized V region fragment (hereinafter referred to as “Diabody”) (Non-patent Literature 28), a disulfide stabilized V region fragment (hereinafter referred to as “dsFv”) (Non-patent Literature 29), or a peptide comprising a complementarity determining region (hereinafter referred to as “CDR”) (Non-patent Literature 30). These antibody fragments are more excellent in transitivity to target tissues than complete antibody molecules (Non-patent Literature 31).